Development of a Method for Evaluating Floor Dry-Cleanability from Wheat Flour in the Food Industry F. Barreca, G.D. Cardinali, E. Borgese, and M. Russo Many productive processes are characterized by inadequate protocols of sanitation that increase the possibility of proliferation of microbial contaminants, especially on surfaces. The use of this method for evaluating the degree of floor cleanability in agri-food companies is important not only to reduce the risk of contamination of products, but also to provide companies with a tool to identify critical issues. The method is based on the usage of bicinchoninic acid assay (BCA) in a solution at a 1:50 ratio of Cu2+ /BCA, which is ideal for detecting the amount of proteins contained in wheat flour residues on industrial flooring. Spectrophotometric analysis allowed identifying maximum absorbance values at 562 nm for different protein concentrations, although the construction of a regression function led to the definition of the intervals of evaluation corresponding to different degrees of cleanliness from residues of wheat flour. The results of the absorbance curves, obtained by applying the proposed evaluation method to 6 tiles commonly used in agri-food buildings, showed the clear persistence of food material on 2 tiles with surface relief. In particular, such tiles showed a higher presence of proteins, with a level of contamination 440% higher. Furthermore, a robotic system was designed to standardize the cleaning method commonly employed in agri-food companies to remove solid particles from flooring. Keywords: agri-food building, cleaning, contamination, floor, food safety Introduction Carpentier 2009). Large amounts of food are often accidentally spilled on floors during processing phases and if not adequately removed may generate a substantial accumulation of substances mainly composed of proteins, starches (Otto and others 2016), and lipids, which are the principal sources of nourishment for the metabolic activities of microorganisms (Donlan and Costerton 2002). In the last few years, the industry of building components has been promoting research on new easy-to-clean materials and surfaces for floors that may meet the market, requirements, and at the same time support environmental sustainability in cleaning and sanitizing food facilities thanks to the limited use of detergents. Standardized tests for the evaluation of floor cleanability, for example, stain resistance and sanitation tests (ISO 10545-14:2015), as well as experimental nonstandardized methods are regularly applied to the most commonly used materials. Schӧler and others (2009) included an optical analysis of the starches found on the surface during the action of a solution of NaOH under pressure; however all of these procedures require the use of detergents and/or disinfectants. Nevertheless, such methods do not allow evaluating the cleanability of surfaces in the specific production sector as they are mainly calibrated for civil buildings. For example, in the case of the stain resistance test, existing regulations provide for a simple visual evaluation of the presence of staining agents, although the agri-food industry requires an evaluation of the presence of traces, even not visible, that may favor the formation of bacterial colonies. The objective of this project was to define a standard method for the evaluation of the dry-cleanability of floors from food residues, in particular, from wheat flours used in the food industry. The method was based on the measurement of protein residues by means of a solution of copper sulfate pentahydrate and bicinJFDS-2016-1765 Submitted 10/26/2016, Accepted 1/19/2017. Authors are with Univ. Mediterranea di Reggio Calabria, Dipto di Agraria, Reggio Calabria, 89100, choninic acid (BCA). A specific application to 6 stoneware tiles with different surface finishing, commonly used in agri-food faItaly. Direct inquiries to author Barreca (E-mail: fbarreca@unirc.it). cilities, was developed to validate the method. Inadequate cleaning of agri-food facilities causes the proliferation of microorganisms, which may be pathogenic. The physicochemical characteristics of food favor the colonization and growth of bacteria and fungi that may even cause life-threatening diseases in human beings (Dzieciol and others 2016). Many productive processes are characterized by inadequate protocols of sanitation that increase the possibility of proliferation of microbial contaminants, especially on surfaces. This results in a considerable impact on public health: every year, in industrialized countries, up to 30% of the population suffer from foodborne diseases (WHO 2007; Bridier and others 2015; Porto and others 2015). Recent studies have confirmed the direct relation between the amount of nutrients found on surfaces after cleaning and initial microbial growth (Kumar and Anand 1998; Blel and others 2010). Low amounts of such substances reduce biofilm formation and make microorganisms vulnerable to disinfectants (Garrett and others 2008). As a result, lower quantities of sanitizing substances are required. It is now common knowledge that 4 main factors affect cleaning: temperature, chemical action, time, and mechanical action (Burkinshaw and others 2011). The last mentioned, which corresponds to the energy spent to remove the residues of food from surfaces, measures the cleanability of a surface and mostly depends on the physico-chemical interactions between the contaminant, the contaminated surface and the cleaning material (Jullien and others 2003; Hoek and Angarwal 2006; Katainen and others 2006). Floors are the surfaces most at risk of contamination, often because of inadequate cleaning protocols that are characterized by low levels of sanitation favoring biofilm formation (Faille and R  C 2017 Institute of Food Technologists doi: 10.1111/1750-3841.13659 Further reproduction without permission is prohibited Vol. 82, Nr. 4, 2017 r Journal of Food Science 939 Food Engineering, Materials Science, & Nanotechnology Abstract: Evaluation of the cleanability of floors . . . Table 1–Composition of the flour contaminating the surface to humidity of no more than 80% and air velocity not exceeding evaluate. 1 m/s. The whole quantity of flour was evenly distributed on the Substances Fats Carbohydrates Sugars Fibers Proteins Sodium chloride Other Quantity (g/100 g wheat flour) surface. 1.10 75.00 0.52 2.50 8.31 0.01 12.56 The 2nd step of the procedure included the cleaning of the surface by scrubbing with a cleaning cloth, a procedure which is usually carried out manually in agri-food facilities before any cleansing and sanitization of surfaces. Proposal of an apparatus to simulate the dry cleaning of the surfaces and sampling of residual wheat flour To simulate correctly the actions performed during manual cleaning, and, above all, to standardize the method and make Materials and Methods it replicable, a special mechanical apparatus was designed (Figure The method included 4 procedural steps: 1). It was composed of a rectangular 100-cm-long and 50-cm-wide • Contamination of a reference surface with a known quantity aluminum frame on which guides were fixed to drive a 13 × 42 cm of flour; sliding block supporting the cleaning cloth. The sliding block was • Cleaning of the surface through a standardized procedure; driven by an electric engine fed by a programmable electronic card • Sampling of the residue of contaminant; • Quantitative evaluation of the residue of contaminant. Food Engineering, Materials Science, & Nanotechnology The material to be evaluated was contaminated by evenly distributing, by means of a powder sprayer, 1 g of type 00 common wheat flour, which is commonly used in the food industry and whose composition is shown in Table 1, on a 20 × 20 cm sample of the flooring surface. The wheat flour was distributed on the analyzed surface in a controlled environment at a temperature of around 25 °C, with 2+ Figure 1–Automated mechanical system for the simulation of the tile Figure 3–Test solution at a 1:50 ratio of Cu : BCA. cleaning. Figure 2–The sliding block. 940 Journal of Food Science r Vol. 82, Nr. 4, 2017 Figure 4–Spectrum curve for a solution at a 1:25 ratio of Cu2+ /BCA containing the 4 different quantities of proteins. Evaluation of the cleanability of floors . . . The cloth used to simulate cleaning operations, which was fixed between the sliding block and the surface to analyze, was obtained from the Kimberly-Clark Professional WYPALL X80 Hydroknit. It is a cloth commonly used in the cleaning of food companies (Figure 2). An important feature of such a mechanical prototype is its transportability, which allows using it not only on samples of tiles in the laboratory but also on site, directly on floors. Furthermore, its programmability favors its adaptation to various production sectors and different methods of cleaning. In order to perform laboratory tests also on limited portions of surfaces, such as single tiles, a specific support was designed to ensure the correct laying of the sample. It was made of a nonporous material and had a size which did not affect the regular movement of the sliding block and allowed firmly fixing the system. The sample of the surface to analyze was lodged in a hole at the center of the slide bed. Moreover, the surface was made planar to the support by a manual leveling system. After cleaning the surface with the automated system, the sample was washed thoroughly with a jet of 100 mL distilled water, which was completely recovered, together with the suspended residues of flour, to perform subsequent analyses. C Evaluation of residual traces of proteins The method of BCA (Brady and Macnaughtan 2015) was used to detect residual flour after cleaning. Such a method allows detecting the presence of proteins. It is commonly used during the monitoring phases of hazard analysis critical control point (HACCP); Figure 5–Spectrum curve for a solution at a 1:50 ratio of Cu2+ /BCA conit is easy to apply and provides a colorimetric evaluation in about taining the 4 different quantities of proteins. 10’. It is based on the chemical reaction between copper sulfate and the proteins found in the solution that reduce Cu2+ to Cu+ in proportion to the quantity of proteins. Cu+ is then chelated by the BCA, which originates the color change of the solution ranging from green to purple-crimson. To establish the optimal ratio between the copper sulfate pentahydrate used in a 4% aqueous solution and the BCA, which is useful to find even small traces of proteins on the surfaces, 4 test solutions, having Cu2+ /BCA ratios equal to 1:25, 1:50, 1:100, and 1:200, respectively, were compared (Biradar and others 2016). Four 4 mL samples were taken from each of the 4 solutions and then 180, 840, 2530, and 9020 mg of flour were introduced, which approximately corresponded to the following quantities of proteins: 15, 70, 210, and 750 µg, respectively. After 10’, 3 mL of test solution were taken from each test Figure 6–Spectrum curve for a solution at a 1:100 ratio of Cu2+ /BCA tube and then introduced in cuvettes for spectrophotometric containing the 4 different quantities of proteins. Figure 7–Spectrum curve for a solution at a 1:200 ratio of Cu2+ /BCA containing the 4 different quantities of proteins. Figure 8–Regression function of absorbance values related to proteins in a solution at a 1:50 ratio of Cu2+ /BCA. Vol. 82, Nr. 4, 2017 r Journal of Food Science 941 Food Engineering, Materials Science, & Nanotechnology that allowed changing its speed and direction as well as the overall number of cleaning cycles (0 1). The pressure exerted by the sliding block was controlled by a digital sensor placed on the interface between the cloth and the analyzed surface in order to obtain a constant value of 269.51 N/m, which corresponded to the average of the pressure measurements taken on the floor sweepers during the usual sweeping of over 20 different operators. The motion of the sliding block was programmed to reverse its direction only once during the cleaning cycle, so as to simulate the operator’s movement during the manual cleaning procedures. Evaluation of the cleanability of floors . . . Table 2–Maximum absorbance values at 562 nm related to Table 3–Visual characteristics of the tile surface. the 4 Cu2+ /BCA solutions containing 4 different quantities of Type Color Texture proteins. Proteins (µg) Cu2+ /BCA ratio 1:25 1:50 1:100 1:200 15 70 210 750 0.06355 0,10898 0.03136 0.02664 0.07393 0,19813 0.12280 0.05322 0.19509 0,47877 0.29531 0.21120 0.60643 1,11529 0.91142 0.46892 A B C D E F Grey Light brown Grey Light brown Light brown Light brown Smooth Smooth Smooth Smooth With cross-shaped elements in relief With lozenge-shaped elements in relief Table 4–Tile roughness values. Food Engineering, Materials Science, & Nanotechnology analysis (Figure 3). Analyses were performed by means of a UV1600PC spectrophotometer, which allowed obtaining the absorbance curves. The following figures show the spectrum curves and the values of maximum absorbance of the 4 test ratios for the 4 quantities of proteins (Figure 4–7 and Table 2). The comparison of the solutions having different Cu2+ /BCA ratios led to the conclusion that the 1:50 ratio was the solution that allowed better discriminating the different quantities of proteins. The regression function (1) was constructed for this solution to relate the quantity of proteins to the absorbance value at a maximum wavelength of 562 nm. m r (x) = a − b exp−c (x) (1) where r(x) was the content of proteins corresponding to the maximum absorbance value x at 562 nm measured through the spectrophotometric analysis, m was the form factor equal to 1.3352, c was the scale factor equal to 0.0003245, a was equal to 1.2367, and b was equal to 1.1414 (Figure 8). Results and Discussions Tile Parameters Ra (µm) Rp (µm) Rq (µm) Rt (µm) Rv (µm) Rz (µm) A B C D E F 1.891 5.118 2.339 13.869 5.142 10.261 2.655 6.663 3.301 21.574 8.937 15.599 1.840 5.563 2.293 13.758 4.587 10.150 2.125 5.750 2.626 14.442 5.402 11.151 6.748 17.478 8.298 46.214 16.072 33.550 6.743 17.291 8.278 46.462 15.717 33.008 tiles E and F, in particular, were characterized by a finish with elements in relief that improved the grip between the floor and the sole. Surface roughness of the analyzed tiles Specific measurements of the roughness characteristics of the surface were taken for the 6 tiles. Standard roughness parameters Ra , Rq , and Rmax (Flint and others 2000; Sedlaček and others 2011), which are commonly used to describe the characteristics of the surfaces, were not enough to define the contact properties of the material, because similar textures could be characterized by different values of roughness parameters. As a result, parameters Rq (root mean square roughness), Rp (maximum height of peaks), Rv (maximum depth of valleys), Rt (maximum peak to valley height), and Rz (10-point height mean; Table 4) were also measured. With a view of validating the method proposed, it was applied to a series of six 20 × 20 cm stoneware tiles (Table 3), commonly used in the food industry (Figure 9). Tiles featured oversized thickness to ensure high shockresistance, because they are commonly used for heavy-traffic floor Application of the BCA method areas. They had a single compact body mass of noble clays sintered The most difficult step of the method proposed was the samat 1250 °C to obtain a material with a high resistance to thermal shocks and chemical attacks. Their surface was slip-resistant and pling of the residue of wheat flour after the standardized cleaning process with the mechanical device. The method adopted was the complete washing of the tiles by means of a 100 mL jet of distilled water, which allowed completely recovering the residual Figure 9–Stoneware tiles analyzed. 942 Journal of Food Science r Vol. 82, Nr. 4, 2017 Figure 10–UV-V is the absorbance curve of the samples of washing water taken after the dry-cleaning of tiles. Evaluation of the cleanability of floors . . . Figure 11–Maximum absorbance at 562 nm and protein concentration of the samples of washing water taken after the dry-cleaning of tiles. Table 5–Maximum absorbance values at 562 nm of the samples Among the tiles having a texture with no element in relief, D of washing water taken after the dry-cleaning of tiles. was characterized by the lowest value of contamination. It showed A B C D E F Absorbance (%) 0.1617 0.0862 0.1207 0.0789 0.9583 0.2179 Concentration (ppm) 0.4118 0.2173 0.2993 0.1958 2.3259 0.5289 Table 6–Evaluation scales of the degree of flooring cleanliness in relation to absorbance values at 562 nm. Degree Clean Doubtful More than doubtful Dirty More than dirty Very dirty Absorbance (%) ࣘ0.129 0.130–0.217 0.218–0.295 0.296–0.646 0.647–0.925 >0.925 flour from the surface without using any detergent that could alter the final evaluation. Then, 100 mL of test solutions with Cu2+ : BCA ratio equal to 1:50 were added into each washing water recovery container. After about 10’, 6 mL of test solution were taken from each washing water container and introduced into a centrifuge to accelerate the sedimentation of the flour particles in the solution, which, if still suspended, could alter the measurement of the absorbance spectrum. Centrifugation was carried out at 172 × g for 2’ at a temperature of 15 °C. Finally, a 3 mL sample was taken from each test solution for the spectrophotometric analysis, which allowed constructing the spectrum curves and the concentration of the analyzed solutions (Figure 10 and Table 5). Following the use of the cleaning apparatus, a visual inspection of the analyzed tiles highlighted the clear persistence of food material on the slip-resistant texture and, in particular, on tile E, due to the surface relief which affected the contact with the cleaning surface and hindered the removal of the residual flour. This observation was confirmed by the results of the absorbance curves obtained by applying the proposed evaluation method to the 6 tiles. Such tiles showed a higher presence of proteins on tile E than on tile F, with a level of contamination 440% higher. Furthermore, tiles E and F showed a high roughness value (Ra > 6 µm), about 3.2 times higher, on average, than that of the tiles without elements in relief. an absorbance equal to about 0.0789% and a protein concentration equal to 0.1958 ppm. It was followed by C, B, and A (Figure 11). Conclusions To show the degree of cleanliness of a surface in a clearer and more immediate manner, the results of the spectrophotometric analysis can be expressed through a qualitative evaluation scale. Regression function (1), which describes the correlation between the maximum absorbance value of the solution measured at 562 nm and the protein concentration, allows defining the intervals of evaluation corresponding to different degrees of cleanliness from residues of protein (Table 6). Such a scale can be easily and quickly applied and could be used during monitoring and control steps or to compare different surfaces. The application of the method proposed in this paper allowed validating its effectiveness and relatively simple use. The importance to have a standardized method of analysis that could be easily applied and allow evaluating the cleanability of surfaces from wheat flour residues and, in particular, of floors in agri-food facilities, plays a crucial role in the search for design solutions for sustainable buildings. For instance, tests showed that the elements in relief on the surfaces of floor areas were an obstacle to cleaning, even if they could improve grip. Actually, the wide range of products and components offered by the market of building materials requires a careful evaluation and analysis of specific uses. A floor could be suitable for a cheese factory but not for a mill, since both the nature of the contaminant and the level of hygiene and safety to guarantee in relation to the specific product are different. The physico-chemical characteristics of the contaminant play a fundamental role in cleaning procedures and, above all, in how its persistence on the surfaces is examined. In this paper, the BCA method was used. It seems to be suitable to detect traces of contaminants containing proteins but it is not equally suitable to highlight the presence of contaminants that do not contain such molecules. Therefore, it is important to continue to search for and establish the most appropriate methods and techniques to detect not only the presence but also low quantities of different types of molecule. Finally, the application of the method highlighted the complex interactions existing between materials, their surface finish and contaminants. Further studies and research can better evaluate the links and correlations between materials and contaminants to direct technical and design choices towards Vol. 82, Nr. 4, 2017 r Journal of Food Science 943 Food Engineering, Materials Science, & Nanotechnology Tile Evaluation of the cleanability of floors . . . more suitable and appropriate materials in relation to the specific food production process. Acknowledgment This paper was developed within the Natl. Operational Programme Research and Competitiveness research project 2007–2013 code PON03PE_00090_1, funded by the European Community. References Food Engineering, Materials Science, & Nanotechnology Biradar AA, Biradar AV, Sun T, Chan Y, Huang X. 2016. Sensors and actuators B: chemical bicinchoninic acid-based colorimetric chemosensor for detection of low concentrations of cyanide. Sens Actuat: B 222:112–19. Blel W, Legentilhomme P, Le Gentil-Lelièvre C, Faille C, Legrand J, Bénézech T. 2010. Cleanability study of complex geometries: interaction between B. cereus spores and the different flow eddies scales. Biochem Eng J 49:40–51. Brady PN, Macnaughtan MA. 2015. Evaluation of colorimetric assays for analyzing reductively methylated proteins: biases and mechanistic insights. Anal Biochem 491:43–51. Bridier A, Sanchez-vizuete P, Guilbaud M, Piard J, Naı̈tali M. 2015. Biofilm-associated persistence of food-borne pathogens. Food Microbiol 45:167–78. Burkinshaw SM, Howroyd J, Kumar N, Kabambe O. 2011. Dyes and pigments The wash-off of dyeings using interstitial water part 2: Bis ( aminochlorotriazine ) reactive dyes on cotton. Dyes Pigms 91(2):134–44. Donlan RM, Costerton JW. 2002. Biofilms: survival mechanisms of clinically relevant microorganisms. Clin Microbiol Rev 15(2):167–93. Dzieciol M, Schornsteiner E, Muhterem-Uyar M, Stessl B, Wagner M, Schmitz-esser S. 2016. International journal of food microbiology bacterial diversity of floor drain bio films and drain waters in a Listeria monocytogenes contaminated food processing environment. Int J Food Microbiol 223:33–40. 944 Journal of Food Science r Vol. 82, Nr. 4, 2017 Faille C, Carpentier B. 2009. Food contact surfaces, surface soiling and biofilm formation. In: Fratamico M, Annous BA, Guenther IV NW, editors. Biofilms in the food and beverage industries. 1st ed. Cambridge, UK: Woodhead Publishing. p. 303–30. Flint SH, Brooks JD, Bremer PJ. 2000. Properties of the stainless steel substrate, influencing the adhesion of thermo-resistant streptococci. J Food Eng 43(4):235–42. Garrett TG, Bhakoo M, Zhang Z. 2008. Bacterial adhesion and biofilms on surfaces. Prog Nat Sci 18:1049–56. Hoek EMV, Agarwal GK. 2006. Extended DLVO interactions between spherical particles and rough surfaces. J Coll Interface Sci 298:50–8. ISO 10545-14:2015.Ceramic tile - Part 14: Determination of resistance to stains. Geneva. Jullien C, Bénézecha T, Carpentier B, Lebret V, Faille C. 2003. Identification of surface characteristics relevant to the hygienic status of stainless steel for the food industry. J Food Engineer 56(1):77–87. Katainen J, Paajanen M, Ahtola E, Pore V, Lahtinen J. 2006. Adhesion as an interplay between particle size and surface roughness. J Coll Interf Sci 304:524–9. Kumar CG, Anand SK. 1998. Significance of microbial biofilms in food industry: a review. Int J Food Microbiol 42(1-2):9–27. Otto C, Zahn S, Hauschild M, Babick F, Rohm H. 2016. Comparative cleaning tests with modified protein and starch residues. J Food Engineer 178:145–50. Porto SMC, Valenti F, Cascone G, Arcidiacono C. 2015. Thermal insulation of a flour mill to improve effectiveness of the heat treatment for insect pest control. CIGR Journal. Special issue 2015. p. 94–104. Schöler M, Fuchs T, Augustin W, Scholl S, Majschak JP. 2009. Monitoring of the local cleaning efficiency of pulsed flow cleaning procedures. Proceedings of 8th International Conference on Heat Exchanger Fouling and Cleaning, 2009 June 14–19, Schladming, Austria. p. 455– 63. Sedlaček M, Silva-Vilhena ML, Podgornik B, Vižintin J. 2011. Surface topography modelling for reduced friction. J Mech Eng 57:674–80. WHO. 2007. Food safety and foodborne illness. Fact Sheet No. 237, World Health Organization.